← Back to Sessions

WITHDRAWN Mutual Information Between Seismic and Geodetic Data Revealed with Machine Learning in Mexico

Description: 

WITHDRAWN Slow-slip events are typically seen in geodetic measurements and can occur during "slow earthquakes" recorded in seismic data. The connection between these two phenomena is still being discussed nowadays, primarily due to the difficulty in identifying the specific seismic signatures of slow earthquakes. Detecting the seismic signatures associated with slow-slip events would be a crucial step toward understanding these phenomena and their role in the seismic cycle. This study explores how utilizing machine learning can aid in understanding the relationship between geodetic and seismic data. Through analyzing a decade of continuous seismograms from a station in Guerrero, Mexico, using a deep scattering network and independant component analsyis, we identify the seismic features that correspond to geodetic movements, suggesting that the seismic signals of slow-slip events can be (empirically) defined. Additionally, we investigate the properties of the associated seismic wavefield to gain insight into how these events contribute to overall displacement seen in geodetic measurements.

Session: Opportunities and Challenges for Machine Learning Applications in Seismology

Type: Oral

Date: 4/19/2023

Presentation Time: 03:00 PM (local time)

Presenting Author: Leonard Seydoux

Student Presenter: No

Invited Presentation: 

Authors